Geospatial analysis as experimental archaeology

In the more than 25 years since Allen et al. (1990), GIS and other kinds of geospatial analysis have become tools used almost as ubiquitously in archaeology as the trowel and the total station. However, can we consider it a "paradigm-shifter?" One fundamental distinction between archaeology and other scientific pursuits is the lack of a formal experimental procedure for testing large-scale hypotheses. We can experiment with some material culture methods or archaeological 'models' on a 1:1 analogue scale, but we rarely examine ideas about larger mechanisms; particularly those that encompass wide geographic areas in a formal experimental way. Geospatial technologies give us new tools and abilities to recognize patterns in archaeological sites and landscapes. Nevertheless, have they truly changed the way we make the transition from material remains to interpreting human behavior? We tend to present geospatial research that is either descriptive or methodological in nature rather than interpretive or explanatory. What is missing is the recognition that the 'patterns' we can see are an incomplete and abstract product of past human agency or behavior that cannot be worked backwards from, but must be envisioned as mechanisms in action. Within a mechanistic framework, we can experiment with archaeological research questions in much greater depth and detail, in a manner more akin to psychology than the 'harder' sciences. Although these techniques bring with them some theoretical assumptions and methodological challenges, their outcomes can provide logical and convincing visualizations of dynamic phenomena in enlightening ways. Presented here are several brief examples. (C) 2017 Elsevier Ltd. All rights reserved.